Högskolan i Skövde

The manufacturing industry is undergoing a profound transformation toward smart, digital, and flexible production systems under the Industry 4.0 framework. Within this paradigm, Digital Twin (DT) serves as a key enabler, bridging physical and digital domains to simulate, analyse, and optimise manufacturing operations. Concurrently, robotic systems, enhanced by smart sensor perception, Industrial Internet of Things connectivity, and adaptive control mechanisms, are increasingly deployed to handle complex and dynamic tasks. However, the evolving demands of the modern manufacturing industry require a high degree of flexibility and responsiveness, necessitating more intelligent solutions. The Robot Digital Twin (RDT) has emerged as a transformative approach, facilitating dynamic adaptation and continuous operational improvement. This review offers a comprehensive examination of the literature on RDT in manufacturing from both technology and application perspectives, aiming to provide insight for researchers and practitioners in Industry 4.0. The paper introduces a four-layer RDT system architecture and summarises how Industry 4.0 technologies, e.g., the Industrial Internet of Things, Cloud/Edge Computing, 5 G, Virtual Reality, Modelling and Simulation, and Artificial Intelligence, converge and influence the RDT system based on this architecture. Furthermore, the review covers domain-specific and system-level applications, such as assembly, machining, grasping, material handling, human-robot interaction, predictive maintenance, and additive manufacturing systems, with an analysis of their development status. Finally, the trends, practical challenges, and future research directions for RDT systems in manufacturing are summarised at different levels.

Dynamic order picking has usually demonstrated significant impacts on production efficiency in warehouse management. In the context of an automotive-part warehouse, this paper addresses a dynamic multi-tour order-picking problem based on a novel attention-aware deep reinforcement learning-based (ADRL) method. The multi-tour represents that one order-picking task must be split into multiple tours due to the cart capacity and the operator’s workload constraints. First, the multi-tour order-picking problem is formulated as a mathematical model, and then reformulated as a Markov decision process. Second, a novel DRL-based method is proposed to solve it effectively. Compared to the existing DRL-based methods, this approach employs multi-head attention to perceive warehouse situations. Additionally, three improvements are proposed to further strengthen the solution quality and generalization, including (1) the extra location representation to align the batch length during training, (2) the dynamic decoding to integrate real-time information of the warehouse environment during inference, and (3) the proximal policy optimization with entropy bonus to facilitate action exploration during training. Finally, comparison experiments based on thousands of order-picking instances from the Swedish warehouse validated that the proposed ADRL could outperform the other twelve DRL-based methods at most by 40.6%, considering the optimization objective. Furthermore, the performance gap between ADRL and seven evolutionary algorithms is controlled within 3%, while ADRL can be hundreds or thousands of times faster than these EAs regarding the solving speed.

Maintenance of assembly and manufacturing equipment is crucial to ensure productivity, product quality, on-time delivery, and a safe working environment. Predictive maintenance is an approach that utilises the condition monitoring data to predict the future machine conditions and makes decisions upon this prediction. The main aim of the present research is to achieve an improvement in predictive condition-based maintenance decision making through a cloud-based approach with usage of wide information content. For the improvement, it is crucial to identify and track not only condition related data but also context data. Context data allows better utilisation of condition monitoring data as well as analysis based on a machine population. The objective of this paper is to outline the first steps of a framework and methodology to handle and process maintenance, production, and factory related data from the first lifecycle phase to the operation and maintenance phase. Initial case study aims to validate the work in the context of real industrial applications.

The presented work is toward population-based predictive maintenance of manufacturing equipment with consideration of the automaticselection of signals and processing methods. This paper describes an analysis performed on double ball-bar measurement from a population ofsimilar machine tools. The analysis is performed after aggregation of information from Computerised Maintenance Management System,Supervisory Control and Data Acquisition, NC-code and Condition Monitoring from a time span of 4 years. Economic evaluation is performedwith use of Monte Carlo simulation based on data from real manufacturing setup.

The cloud technology provides sustainable solutions to the modern industrial robotic cells. Within the context, the objective of this research is to minimise the energy consumption of robots during assembly in a cloud environment. Given a robot path and based on the inverse kinematics and dynamics of the robot from the cloud, a set of feasible configurations of the robot can be derived, followed by calculating the desirable forces and torques on the joints and links of the robot. Energy consumption is then calculated for each feasible configuration along the path. The ones with the lowest energy consumption are chosen. Since the energy-efficient robot configurations lead to reduced overall energy consumption, this approach becomes instrumental and can be applied to energy-efficient robotic assembly. This cloud-based energy-efficient approach for robotic applications can largely enhance the current practice as demonstrated by the results of three case studies, leading towards sustainable manufacturing.

The ability to adaptively control manufacturing equipment in cloud environments is becoming increasingly more important. Industry 4.0, supported by Cyber Physical Systems and the concept of on-demand, scalable and pay-for-usage resource-sharing in cloud environments offers many promises regarding effective and flexible manufacturing. For implementing the concept of manufacturing services in a cloud environment, a cloud control approach for the sharing and control of networked manufacturing resources is required. This paper presents a cloud service-based control approach which has a product perspective and builds on the combination of event-driven IEC 61499 Function Blocks and product manufacturing features. Distributed control is realised through the use of a networked control structure of such Function Blocks as decision modules, enabling an adaptive run-time behaviour. The control approach has been developed and implemented as prototype systems for both local and distributed manufacturing scenarios, in both real and virtual applications. An application scenario is presented to demonstrate the applicability of the control approach. In this scenario, Assembly Feature-Function Blocks for adaptive control of robotic assembly tasks have been used.

Augmented Reality (AR) has shown its potential in supporting operators in manufacturing. AR-glasses as a platform both in industrial use are emerging markets, thereby making portable and hands-free AR more and more feasible. An important aspect of integrating AR as a support tool for operators is their acceptance of the technology. This paper presents the results of interviewing operators regarding their view on AR technology in their field and observing them working in automotive engine assembly and how they interact with current instructions. The observations and follow-up questions identified three main aspects of the information that the operators looked at: validating screw torque, their current assembly time, and if something went wrong. The interviews showed that a large amount of the operators were positive towards using AR in assembly. This has given an insight in both the current information interaction the operators do and their view on the potential in using AR. Based on these insights we suggest a mock-up design of an AR-interface for engine assembly to serve as a base for future prototype designs.

In sustainable manufacturing, the proper maintenance is crucial to minimise the negative environmental impact. In the context of Cloud Manufacturing, Internet of Things and Big Data, amount of available information is not an issue, the problem is to obtain the relevant information and process them in a useful way. In this paper a maintenance decision support system is presented that utilises information from multiple sources and of a different kind. The key elements of the proposed approach are processing and machine learning method evaluation and selection, as well as estimation of long-term key performance indicators (KPIs) such as a ratio of unplanned breakdowns or a cost of maintenance approach. Presented framework is applied to machine tool linear axes. Statistical models of failures and Condition Based Maintenance (CBM) are built based on data from a population of 29 similar machines from the period of over 4 years and with use of proposed processing approach. Those models are used in simulation to estimate the long-term effect on selected KPIs for different strategies. Simple CBM approach allows, in the considered case, a cost reduction of 40% with the number of breakdowns reduced 6 times in respect to an optimal time-based approach.

In this paper, knowledge transfer is defined as a process of disseminating both technological and theoretical understanding as well as enhancing both industrial and academic knowledge through conducted research to project partners collaborating within a research project. To achieve this, a new methodology called 'user groups' is introduced. It facilitates knowledge transfer between project participants in collaborative research programs engaging both experienced and unexperienced partners regardless of level of input. The introduced methodology 'user groups' provides tools for collaborating with several research partners even though their levels of engagement in the project and prior research experience may vary without dividing them into separate groups. It enables all project partners to gain new knowledge and by so doing extending the knowledge society. The case study shows that the eight engaged companies are able to cooperate, achieve their own objectives and, both jointly and individually, contribute to the overall project goals.

This paper presents a novel system using Augmented Reality and Expert Systems to enhance the quality and efficiency of shop-floor operators. The novel system proposed provides an adaptive tool that facilitates and enhances support on the shop-floor, due to its ability to dynamically customize the instructions displayed, dependent upon the competence of the user. A comparative study has been made between an existing method of quality control instructions at a machining line in an automotive engine plant and this novel system. It has been shown that the new approach outcompetes the existing system, not only in terms of perceived usability but also with respect to two other important shop-floor variables: quality and productivity. Along with previous research, the outcomes of these test cases indicate the value of using Augmented Reality technology to enhance shop-floor operators’ ability to learn and master new tasks.